Method of determining the direction of arrival of a radio signal, as well as radio base station and radiocommunications system

Abstract

The determination of the direction of arrival of a radio signal by an antenna array connected to a base station of a radiocommunications system, particularly of an SDMA (Space Division Multiple Access) mobile radio system, is complicated by multipath. A method is known which uses the so-called ESPRIT algorithm and which is especially suited for a reliable estimation of direction, since the receive level (Sm) and the phase position (φm) are measured for each radiating element and entered in a symmetrical matrix (A) in order to then determine the direction of arrival by eigenvalue decomposition. A simpler and faster method (100) is proposed which involves computing the eigenvector (Wn) corresponding to the dominant eigenvalue (λ1) of this matrix (A), which indicates the direction of arrival (DOA) of the strongest radio signal (S) (steps 130 and 140). The computationally intensive eigenvalue decomposition is eliminated.

We claim: 
     
       1. A method of determining a direction of arrival of a radio signal distorted by multipath and being received by an antenna array having at least two radiating elements, said method comprising the steps of: collecting data for each of said at least two radiating elements, wherein said data specify a receive level and a phase position;   entering said data in a matrix (A) having M columns and M rows; and   computing a required eigenvector that corresponds to a dominant eigenvalue of said matrix (A), wherein said required eigenvector is computed in the following part steps: a) setting a counter (i) to 1 (i=1) and estimating a first vector (W i , with i=1), which is to correspond to said required eigenvector, for said dominant eigenvalue of said matrix (A); and   b) computing an ith product vector (A·W i ) from said matrix (A) and said first vector (W i ), and computing a Rayleigh quotient (RQ) from the formula:   RQ=(A·W.sub.i ·W.sub.i)/(W.sub.i ·W.sub.i); and         pointing a radiation pattern of said antenna array by evaluating a last computed vector (W i=n ), which is said required eigenvector, wherein said required eigenvector indicates said direction of arrival of a strongest radio signal.   
     
     
       2. A method as claimed in claim 1, wherein said first vector, which is to correspond to said required eigenvector, is determined by reading contents of a first column of said matrix (A). 
     
     
       3. A method as claimed in claim 1, wherein the step of computing a required eigenvector further comprises the part step of: c) determining a maximum (d) of the components of said product vector (A·W i ), and computing a new vector (W i+1 ) by weighting said product vector (A·W i ) with the inverse of said maximum (i.e., W i+1  =(1/d)·A·W i ).   
     
     
       4. A method as claimed in claim 3, wherein said part steps b) and c) are iteratively repeated until said Rayleigh quotient changes by less than a predeterminable tolerance value. 
     
     
       5. A method as claimed in claim 3, wherein said part steps b) and c) are iteratively repeated until a predetermined number of cycles is reached. 
     
     
       6. A method as claimed in claim 1, further comprising the step of evaluating said last computed vector to determine a direction of transmission by weighting said last computed vector with a weight factor, wherein said weight factor is determined from an offset between a transmitted frequency and a received frequency. 
     
     
       7. A radio base station comprising: an antenna array having at least two radiating elements for receiving a radio signal distorted by multipath;   an evaluating device connected to said antenna array, wherein said evaluating device collects data for each of said at least two radiating elements, and wherein said data specify a receive level and a phase position for determining a direction of arrival of said radio signal; and   a memory device, connected to said evaluating device, for storing said data in the form of a symmetrical matrix (A) having M columns and M rows;   wherein said evaluating device computes an eigenvector corresponding to a dominant eigenvalue of said symmetrical matrix (A) in the following manner: a) setting a counter (i) to 1 (i=1) and estimating a first vector (W i , with i=1), which is to correspond to said eigenvector, for said dominant eigenvalue of said matrix (A); and   b) computing an ith product vector (A·W i ) from said matrix (A) and said first vector (W i ), and computing a Rayleigh quotient (RQ) from the formula:   RQ=(A·W.sub.i ·W.sub.i)/(W.sub.i ·W.sub.i);         wherein said radio base station points a radiation pattern of said antenna array by evaluating a last computed vector (W i=n ), which is said eigenvector, and wherein said eigenvector indicates said direction of arrival of a strongest radio signal.   
     
     
       8. A radio base station as claimed in claim 7, wherein said evaluating device computes said eigenvector by further determining a maximum (d) of the components of said product vector (A·W i ), and computing a new vector (W i+1 ) by weighting said product vector (A·W i ) with the inverse of said maximum (i.e., W i+1  =(1/d)·A·W i ). 
     
     
       9. A radiocommunications system comprising at least one radio base station, wherein said at least one radio base station comprises: an antenna array having at least two radiating elements for receiving a radio signal distorted by multipath;   an evaluating device connected to said antenna array, wherein said evaluating device collects data for each of said at least two radiating elements, and wherein said data specify a receive level and a phase position for determining a direction of arrival of said radio signal; and   a memory device, connected to said evaluating device, for storing said data in the form of a symmetrical matrix (A) having M columns and M rows;   wherein said evaluating device computes an eigenvector corresponding to a dominant eigenvalue of said symmetrical matrix (A) in the following manner: a) setting a counter (i) to 1(i=1) and estimating a first vector (W i , with i=1), which is to correspond to said eigenvector, for said dominant eigenvalue of said matrix (A); and   b) computing an ith product vector (A·W i ) from said matrix (A) and said first vector (W i ), and computing a Rayleigh quotient (RQ) from the formula:   RQ=(A·W.sub.i ·W.sub.i)/(W.sub.i ·W.sub.i);         wherein said at least one radio base station points a radiation pattern of said antenna array by evaluating a last computed vector (W i=1 ), which is said eigenvector, and wherein said eigenvector indicates said direction of arrival of a strongest radio signal.   
     
     
       10. A radiocommunications system as claimed in claim 9, wherein said evaluating device computes said eigenvector by further determining a maximum (d) of the components of said product vector (A·W i ), and computing a new vector (W i+1 ) by weighting said product vector (A·W i ) with the inverse of said maximum (i.e., W i+1  =(1/d)·A·W i ).